Aggressive behaviour at regrouping is a poor predictor of chronic aggression in stable social groups
Publication Type: |
Journal Article |
Year of Publication: |
2017 |
Authors: |
Simon P. Turner, Ian M. Nevison, Suzanne Desire, Irene Camerlink, Rainer Roehe, Sarah H. Ison, Marianne Farish, Mhairi C. Jack, Richard B. D’Eath |
Publication/Journal: |
Applied Animal Behaviour Science |
Keywords: |
aggression, cluster analysis, fighting, lesion, pig, social |
ISBN: |
0168-1591 |
Abstract:
Commercial pigs globally are routinely mixed into new social groups. This results in regrouping aggression predominantly during the first 24 h which compromises welfare and productivity. Chronic aggression persists thereafter and is also undesirable. Management strategies are needed that reduce the costs of aggression in both of these contexts. Pigs vary greatly in aggressive behaviour and numbers of skin lesions. This study examined how regrouping behaviour affects immediate and long-term lesion counts with a specific focus on understanding the behaviour of pigs with few lesions in both social contexts. Aggressive behaviour from 1163 growing pigs was observed for 24 h post-regrouping and fresh lesions were counted 24 h and 3 weeks post-regrouping. Similarity between pigs was calculated using all behavioural traits recorded during the 24 h post-regrouping. Clusters of pigs were formed using furthest neighbour clustering with a stopping rule of 80% similarity. Five clusters of pigs representing 90% of the population (1047 pigs) were identified. For each regrouping aggressive behaviour trait and for fresh lesion counts 24 h post-regrouping the means differed significantly (P < 0.0001) between clusters. The most extreme clusters were characterised by extremely high or low levels of aggression with the other three clusters characterised by pigs that were unaggressive losers, selectively aggressive or with long fights. Statistically significant (P < 0.05–P < 0.001) but numerically small differences between clusters were found in lesion count 3 weeks post-regrouping. Pigs were separately categorised based upon their combination of lesion counts recorded 24 h and 3 weeks post-regrouping. Pigs showing similar behaviour at regrouping displayed wide ranging combinations of acute and chronic lesion outcomes. Pigs with particularly low lesion counts at both regrouping and 3 weeks post-regrouping were found in all 5 clusters. Avoidance of aggressive behaviour at regrouping resulted in few lesions at 24 h but more lesions at 3 weeks. Increasing the proportion of pigs in the population that receive few lesions from both regrouping and chronic aggression may require management strategies that manipulate behaviour in both contexts. Long-term costs of avoiding regrouping aggression, represented by lesion counts three weeks after re-grouping, show that regrouping aggression may retain an important function in domesticated pigs and potentially in other species.